Triple

T10926364
Position Surface form Disambiguated ID Type / Status
Subject Osmosis Jones E258078 entity
Predicate writer P1360 FINISHED
Object Marc Hyman
Marc Hyman is an American screenwriter best known for co-writing the live-action/animated comedy film "Osmosis Jones."
E893532 NE FINISHED

How this triple was built (4 steps)

Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.

NER Named-entity recognition gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: Marc Hyman | Statement: [Osmosis Jones, writer, Marc Hyman]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Marc Hyman
Context triple: [Osmosis Jones, writer, Marc Hyman]
  • A. Marc Rosen
    Marc Rosen is an American businessman and talent agent best known as the younger husband of classic Hollywood actress Arlene Dahl.
  • B. Mark Rosner
    Mark Rosner is a screenwriter best known for co-writing the 1996 action film "The Rock."
  • C. Mark Rosman
    Mark Rosman is an American film and television director and screenwriter best known for his work on family and teen-oriented movies and series, including projects for Disney.
  • D. Jay Landsman
    Jay Landsman is a Baltimore police sergeant and homicide supervisor best known as a character in the television series "The Wire," inspired by and partly portrayed by the real-life Baltimore detective of the same name.
  • E. Michael Haussman
    Michael Haussman is an American director and filmmaker best known for his work on high-profile music videos and commercials.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg Description generation gpt-5.1
Instruction
Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: Marc Hyman
Triple: [Osmosis Jones, writer, Marc Hyman]
Generated description
Marc Hyman is an American screenwriter best known for co-writing the live-action/animated comedy film "Osmosis Jones."
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Marc Hyman
Target entity description: Marc Hyman is an American screenwriter best known for co-writing the live-action/animated comedy film "Osmosis Jones."
  • A. Marc Rosen
    Marc Rosen is an American businessman and talent agent best known as the younger husband of classic Hollywood actress Arlene Dahl.
  • B. Mark Rosner
    Mark Rosner is a screenwriter best known for co-writing the 1996 action film "The Rock."
  • C. Mark Rosman
    Mark Rosman is an American film and television director and screenwriter best known for his work on family and teen-oriented movies and series, including projects for Disney.
  • D. Jay Landsman
    Jay Landsman is a Baltimore police sergeant and homicide supervisor best known as a character in the television series "The Wire," inspired by and partly portrayed by the real-life Baltimore detective of the same name.
  • E. Michael Haussman
    Michael Haussman is an American director and filmmaker best known for his work on high-profile music videos and commercials.
  • F. None of above. chosen

Provenance (5 batches)

The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.

Step Stage Batch ID Status When
creating Elicitation batch_69d6aa864ed88190818280ab6791d065 completed April 8, 2026, 7:20 p.m.
NER Named-entity recognition batch_69d7709165188190aa30dd08deddade4 completed April 9, 2026, 9:25 a.m.
NED1 Entity disambiguation (via context triple) batch_69e217369b648190914c58db6f6e0200 completed April 17, 2026, 11:19 a.m.
NEDg Description generation batch_69e21d8aea2881908ac8f5225b8739c5 completed April 17, 2026, 11:46 a.m.
NED2 Entity disambiguation (via description) batch_69e21eb18a1881908ded331db89063ed completed April 17, 2026, 11:51 a.m.
Created at: April 8, 2026, 9:22 p.m.